----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/38324/#review99684 -----------------------------------------------------------
Ship it! include/mesos/master/allocator.hpp (line 162) <https://reviews.apache.org/r/38324/#comment156652> Why don't we default to None here? src/master/allocator/mesos/hierarchical.hpp (line 220) <https://reviews.apache.org/r/38324/#comment156657> See comment below. src/master/allocator/mesos/hierarchical.hpp (line 402) <https://reviews.apache.org/r/38324/#comment156653> We don't need the slave id anymore. src/master/allocator/mesos/hierarchical.hpp (line 403) <https://reviews.apache.org/r/38324/#comment156656> Since this is a specific implementation of an allocator, and the only time we send out inverse offers is at the slave level for maintenance primitives, I think we don't need to capture the unavailableResources for now. We can leave a comment here explaining this. src/master/allocator/mesos/hierarchical.hpp (line 411) <https://reviews.apache.org/r/38324/#comment156654> same here and in the implementation. src/master/allocator/mesos/hierarchical.hpp (line 423) <https://reviews.apache.org/r/38324/#comment156655> We don't need to compare the slaveId anymore. src/master/allocator/mesos/hierarchical.hpp (line 589) <https://reviews.apache.org/r/38324/#comment156664> I think we need to clear inverse offer filters as well? src/master/allocator/mesos/hierarchical.hpp (lines 908 - 909) <https://reviews.apache.org/r/38324/#comment156663> Let's clear the inverse offer filters explicitly to force frameworks to reassess their calculations. src/master/allocator/mesos/hierarchical.hpp (line 1130) <https://reviews.apache.org/r/38324/#comment156665> I think we need to clear inverse offers filters as well? src/master/allocator/mesos/hierarchical.hpp (line 1331) <https://reviews.apache.org/r/38324/#comment156658> Adjusted to not include the unavailableResources. src/master/allocator/mesos/hierarchical.hpp (lines 1331 - 1334) <https://reviews.apache.org/r/38324/#comment156660> we can move this above now that we don't need to construct unavailableResources to filter. src/master/allocator/mesos/hierarchical.hpp (line 1335) <https://reviews.apache.org/r/38324/#comment156659> new line src/master/allocator/mesos/hierarchical.hpp (lines 1401 - 1402) <https://reviews.apache.org/r/38324/#comment156662> wrapping - Joris Van Remoortere On Sept. 18, 2015, 10:55 p.m., Artem Harutyunyan wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/38324/ > ----------------------------------------------------------- > > (Updated Sept. 18, 2015, 10:55 p.m.) > > > Review request for mesos, Joris Van Remoortere and Joseph Wu. > > > Bugs: MESOS-3346 > https://issues.apache.org/jira/browse/MESOS-3346 > > > Repository: mesos > > > Description > ------- > > See summary. > > > Diffs > ----- > > include/mesos/master/allocator.hpp fb09e2a6502bc8c78ddcc8a595bcd9320da136ea > src/master/allocator/mesos/allocator.hpp > 171548b2017a0b97124f052c21345668e274d117 > src/master/allocator/mesos/hierarchical.hpp > 3374d63b8311cf10b3108f56b7b167c12a9d7a37 > src/master/master.cpp 1c4e7af7448c05f54c1068d6496ed21c8c359ac5 > src/tests/mesos.hpp 3db97aca921c9216d90384e1eb17030849516454 > > Diff: https://reviews.apache.org/r/38324/diff/ > > > Testing > ------- > > Test in the next review. > > > Thanks, > > Artem Harutyunyan > >
